January 2, 2010 13:56 WSPC/MEDJ
00012
Middle East Development Journal, Vol. 1, No. 2 (2009) 255–273
c Economic Research Forum
CORPORATE CAPITAL STRUCTURE CHOICES IN MENA:
EMPIRICAL EVIDENCE FROM NON-LISTED
FIRMS IN MOROCCO
LAHCEN ACHY
Institut National de Statistique et d’Economie Appliquée (INSEA)
BP 6217 Rabat Institutes, Rabat, Morocco
[email protected]
Received 23 April 2009
Revised 28 August 2009
Based on their perceptions, more than three quarters of Moroccan manufacturing firms
have identified access to finance as one of the major constraints affecting their performance. However, compared to a number of emerging countries, Moroccan firms appear
relatively undercapitalized and more reliant on external finance. These two findings seem
contradictory and have very different policy implications. The purpose of this paper is
to provide a rigorous understanding of the rationale behind financial choices made by
Moroccan firms, and assess the severity of financial constraints they effectively face. The
paper uses a panel dataset covering 550 non-listed manufacturing firms over the period
1998–2003 and investigates both long-term and short-term measures of leverage with the
objective of understanding the factors that shape “debt-equity choice” as well as “debt
maturity structure”.
Our analysis reveals the existence of a negative relationship between asset tangibility
and both aggregate leverage and short-term debt ratio. However, no clear cut relationship
between asset tangibility and long-term debt is uncovered. Small firms tend to increase
their debt instead of opening their capital to outside investors and larger firms seem
to rely much more on their retained earnings for their long-term financial needs. For
short-term debt, size does not appear to matter. The impact of growth is positive on
short-term leverage and irrelevant for long-term leverage. Finally, profitability exerts a
positive effect on long-term leverage and a negative one on short-term leverage.
Keywords: Financial structure; debt-equity choice; debt maturity issue; manufacturing
firms.
1. Introduction
The purpose of this paper is to empirically investigate the determinants of financial
structure in non-listed firms in a developing country, namely Morocco, using a
panel data approach. The paper investigates long-term and short-term measures
of leverage with the objective of understanding the factors behind shaping “debtequity choice” as well as “debt maturity structure” in the context of non-listed
manufacturing firms. The issue is of high relevance for both academic research and
policy-making.
255
January 2, 2010 13:56 WSPC/MEDJ
256
00012
L. Achy
From an academic point of view, there is a relatively vast body of theoryderived literature relating corporate capital structure to firm and industry characteristics. However, most studies use data on listed companies, and frequently focus
on developed countries. Only a limited amount of research has focused on developing countries largely because of data constraints and therefore there is an important
knowledge gap that needs to be tackled.
Moreover, there are valid reasons to expect that capital structure decisions of
non-listed firms are shaped by different factors compared to listed firms. This is particularly the case in developing countries where capital markets are less developed,
the range of financial instruments available to non-listed firms is relatively narrow,
and the lack of rigorous accounting standards and audit controls create higher information asymmetry among stakeholders (Cobham and Subramaniam 1998). Using
firm level data on business environment across 80 countries, Ayyagari et al. (2005)
show that access to finance is the most robust factor that has a direct impact on
firms’ growth. Various other studies suggest that financing obstacles faced by small
and medium-sized enterprises translate into slower growth (Beck et al. 2005, Beck
and Demirguç-Kunt 2006).
From a policy-making point of view, the paper is expected to provide a more
rigorous understanding of the rationale behind financial choices made by Moroccan
manufacturing firms, and assess the severity of financial constraints they effectively
face. Based on their perceptions, more than three-quarters of Moroccan manufacturing firms have identified access to finance as one of the major constraints affecting
their performance (ICA surveya 2004). However, a large cross sectional survey conducted in 48 countries (Beck et al. 2008) reveals that compared to firms in emerging
countries such as Turkey, Egypt, Mexico, Jordan, Poland and Argentina, Moroccan
firms appear relatively undercapitalized and more reliant on external finance.
Thanks to a panel dataset covering some 550 firms over the period 1998–2003,
we are able to extend the existing empirical work on the determinants of financial
structure to non-listed manufacturing firms in the specific context of a developing
MENA country, namely Morocco. In addition, we investigate both long-term and
short-term measures of leverage with the objective of understanding the factors
that shape “debt-equity choice” as well as “debt maturity structure”.
The rest of the paper proceeds as follows. Section 2 examines the main findings
of previous empirical work on leverage determinants and their relevance for nonlisted firms. Section 3 describes data sources and analyzes typical balance sheets
of non-listed firms in the Moroccan manufacturing sector. Section 4 presents the
econometric approach, and discusses the main empirical results. It also compares
and contrasts long-term and short-term determinants of leverage behavior. Finally,
Sec. 5 sums up the main findings and draws their policy implications.
a ICA
stands for Investment Climate Assessment, which is a firm level survey implemented in
collaboration with the World Bank. ICA is a broad survey that covers firm’s activity, organization,
its inputs and outputs, infrastructure and services, finance, labor and training, regulation and
conflict resolution.
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
257
2. Literature Review
The theoretical background underpinning the literature on corporate capital structure dates back to Modigliani and Miller (1958) for whom the value of a company is
independent of its capital structure in a perfect capital market. Therefore, the issue
of optimal capital structure is irrelevant. Since then, economists have developed a
number of theories to explain variation in debt ratios across firms by accounting
for the implications of capital market imperfections. From the perspective of nonlisted companies, Modigliani and Miller’s argument does not seem appropriate for
at least three reasons. First, non-listed companies have access to a limited range of
financial instruments to adjust their current leverage to their desired level. Second,
non-listed firms are generally held by a single owner or by members of the same
family and likely to be reluctant to open their business to outsiders. Third, even
if they decide to open their capital, it would not be easy to convince potential
investors to acquire a stake in unlisted family businesses.b
The agency cost model initiated by Jensen and Meckling (1976) postulates that
shareholder-manger conflict plays a key role in shaping capital structure decisions.
Managers have different incentives in comparison with owners, which may lead them
to waste the firm’s free cash flow. Therefore, the advantage of debt is to reduce free
cash flow available to managers. However, the agency cost model does not really
fit with the situation of non-listed firms. Owners of this category of firms are very
likely to be also involved in their management,c and hence the shareholder-manager
conflict is less relevant to explain their financial structure.
The trade-off theory suggests that firms target an optimal level of leverage to
balance the benefits and costs of debt financing. The main benefit of debt is its
tax deductibility by firms. It therefore pays to borrow as long as a firm has taxable profits. However, the costs of financial distress (Myers 2001) impose limits on
the optimal level of debt targeted by a firm. The costs can be direct out-of-pocket
expenses or indirect such as the reluctance of suppliers to deal with firms in financial distress. This aspect is crucial for non-listed firms for which credit extended
by suppliers generally represents a key source of short-term finance. Furthermore,
trade-off theory fails to explain why firms with higher profitability are often characterized by low debt levels (Rajan and Zingales 1995).
The pecking order model developed by Myers and Majluf (1984) provides such
an explanation. It argues that profitable firms, which generate sufficient cash flows
to meet their capital needs, use their retained earnings first. They opt for debt
as their second choice. Additional equity finance is used as a source of last resort.
Holmes and Kent (1991) argue that leverage decisions of non-listed firms tend to be
shaped by the pecking order theory for two reasons. First, their access to appropriate
b We
would like to thank an anonymous referee for suggesting this caveat of the Modigliani and
Miller’s paradigm in the specific context of non-listed family firms.
c For instance, our data on non-listed Moroccan manufacturing firms indicate that in 89% of cases,
there is no separation between ownership and control.
January 2, 2010 13:56 WSPC/MEDJ
258
00012
L. Achy
external sources of capital is limited. Second, their cost of capital is generally high
due to asymmetrical information in credit markets. In order to circumvent these
constraints, non-listed firms often tend to rely primarily on their internal sources
of funding.
From an empirical point of view, there is a relatively vast body of theory-derived
literature relating corporate capital structure to firm and industry characteristics.
However, most studies use data on listed companies, and frequently focus on developed countries. For example Titman and Wessels (1988) focused on American firms,
Rajan and Zingales (1995) and Aggrawal and Jamdee (2003) investigated the determinants of capital structure in the G-7 countries.
On the other hand, very few papers have dealt with the same issue in developing
countries. Glen and Pinto (1994) found that unlike firms in G7 countries, firms in
developing countries rely more substantially on externally generated funds. Booth
et al. (2001) examined capital structure in 10 developing countries, two of which
were from MENA (Jordan and Turkey).d Their findings indicated that, overall,
capital structure choices in developing countries are affected by the same variables
as in developed countries. Finally, Agarwal and Mohtadi (2004) focused on the
impact of financial sector size and structure of debt-equity ratios in 21 developing
countries. However, in all these cases, only listed firms were covered.
3. Data and Descriptive Analysis
3.1. Data sources
Financial variables used in the paper are expressed in book values and drawn from
balance sheets and income statements of a sample made from 550 Moroccan manufacturing firms over the period 1998–2003. It is important to mention that the same
firms are tracked over the five-year period. To obtain this sample we had to merge
two separate but consistent databases.e The first relies on data collected in 2000
under FACS (Firm Analysis and Competitiveness Survey). The second is based on
data collected in 2004 under ICA (Investment Climate Survey). Both surveys are
jointly conducted by the Moroccan Ministry of Industry and the World Bank.
To our knowledge, this is the most comprehensive dataset on financial variables
that has ever been constructed on the manufacturing sector in Morocco. Nevertheless, we checked the data for consistency to detect potential outliers on the basis of
accounting rules and logical relationships.f
d Countries
covered by Booth et al. (2001) included: Brazil, Mexico, India, South Korea, Jordan,
Malaysia, Pakistan, Thailand, Turkey and Zimbabwe.
e Each firm has a unique code used in both databases.
f We checked for consistency within each accounting period using accounting rules such as (total
assets = total debt + total equity). We also checked for consistency from one period to the other
by looking at the annual changes on aggregates, such as output, employment, and total sales; also
available from the annual manufacturing survey data. We had to remove some 30 firms from our
database due to their numbers that were very unlikely. Overall, and from a pure statistical point
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
259
3.2. Descriptive analysis
The purpose of this section is to provide a brief descriptive analysis of sources of
funds among manufacturing firms in Morocco. Although more than thirty items are
included in the database, we perform a certain amount of aggregation in order to
focus on the respective importance of the main components. Data are averaged to
provide the liability side of the balance sheet for a hypothetical firm with the mean
characteristics of the sample. The results are presented in Table 1 for the first and
the last available years in our dataset, namely 1998 and 2003. The last column of
Table 1 reports, for each component, the average over the six-year period.
The share of equity, which represents the amount of money initially invested
by owners plus any retained earnings, stands at 37% over the period under study.
It increases from around 32% in 1998 to more than 40% in 2004 due mainly to
the drastic jump recorded in retained earnings. The share of the latter in total
financing more than tripled in six years, going up from 4.9% in 1998 to 16.7%
in 2003. Retained earning is an accumulated number, with each year’s retained
earnings being added to the amount from prior years. Although it is natural for
firms to have a larger pool of retained earnings relative to total equity as they
become older, the increase recorded in our case is also explained by the decision
in 1997 by the government to exempt up to 20% of profits retained for investment
purposes. Thus, Moroccan firms appear to be sensitive to fiscal incentives in their
trade-offs between equity and debt.
Debt, on the other hand, seems to play a central role in the capital structure
of the Moroccan manufacturing sector. This is particularly the case for short-term
debt (current liabilities), which accounts for roughly the equivalent of half of total
Table 1. Financial structure of Moroccan non-listed manufacturing firms.
Total equity
Capital
Retained earnings
Long-term debt
Short-term debt
Accounts payable
Credit to suppliers
Other short-term creditors
Short-term bank loans
Total liabilities and equity
1998
2003
Average
1998–2003
31.9
27.0
4.9
13.0
55.2
37.0
21.1
15.9
18.2
100
40.4
23.7
16.7
13.6
46.0
35.1
23.1
12.0
10.9
100
37.0
25.0
12.0
14.1
48.9
33.4
18.9
14.5
15.5
100
Source: Moroccan FACS (Firm Analysis and Competitiveness
Survey) and ICA (Investment Climate Assessment) carried
out respectively in 2000 and 2004.
view, our dataset seems to be of good quality. The issue of credibility of accounting reporting goes
beyond the scope of this paper.
January 2, 2010 13:56 WSPC/MEDJ
260
00012
L. Achy
firms’ assets over the period 1998–2003. Long-term debt, on the other hand, appears
to be of limited contribution as its share does not exceed 14%.
But how does the financial structure of Moroccan firms compare with their
counterparts in other countries?
Booth et al. (2001) reported debt ratios for 17 countries (10 developing and
7 developed countries). For developed countries, debt ratios ranged between 54%
(UK) and 73% (Germany). For developing countries, debt ratios were globally lower
and ranged between 30% in Brazil and a surprising 73% in South Korea. Regarding long-term debt, it appears that its share in developed countries is much more
important compared to short-term debt; it varied between 28% in the UK and 53%
in Japan. Conversely, in developing countries short-term debt exceeded long-term
with the exception of South Korea and India. Although these results are useful for
comparative purposes, their relevance is limited for two reasons. First, they only
cover listed firms. Second, they are not limited to the manufacturing sector.
Thanks to the database constructed by the World Bank through a major cross
sectional survey conducted in 48 countries on 3000 firms, among which 80% are
small and medium (Beck et al. 2008), we are able to perform more appropriate
comparisons. On average external finance accounts for 41% over the entire sample
with minor differences between developed and developing countries. Furthermore,
compared to firms in emerging countries such as Turkey, Egypt, Mexico, Jordan,
Poland and Argentina, Moroccan firms appear relatively undercapitalized and more
reliant on external finance.
Moroccan firms, with a debt ratio amounting to 63%, seem to be located in
the highest debt group with respect to the rest of the countries in the World Bank
sample. However, based exclusively on their long-term debt ratio, Moroccan firms
fall into the lowest debt group. This situation may be a potential source of financial
fragility of Moroccan firms.
One of the basic rules in financial management is that firms need to match the
maturity of their debt with the degree of liquidity of their assets. Therefore, one
simple way to assess financial fragility of Moroccan firms is to test the extent to
which firms comply with this matching rule. At the aggregate level as well as by
size class,g there seems to be no mismatching issue and thus no special worry about
the financial structure of Moroccan firms. Their long-term sources of finance cover
their fixed assets and their current assets are larger than their current liabilities.
When we examine the items under short-term debt, we see that a large proportion is made by credit to suppliers and short-term creditors. The share of short-term
bank loans is relatively limited. Trade credit is usually not free, as suppliers consider all costs including the cost of extending trade credit when setting their prices.
The ICA survey indicates that in case of cash payments, firms benefit from price
discounts. Theoretically, the role of credit extended by suppliers in a firm’s financial
structure depends on how quickly the firm can pay off new balances (Danielson and
g We
constructed three classes on the basis of the number of employees: small firms (less than 50),
medium-sized firms (between 51 and 200) and large firms (more than 200).
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
261
Scott 2004). If a firm cannot make timely payments, probably because it faces liquidity constraints or because additional bank loans are not available, trade credit
can stand as a substitute for bank loans. Otherwise, credit from suppliers can be
seen as complementary to bank loans.
On the basis of the ICA survey, credit granted by suppliers represents almost
one fifth of total assets and has an average duration of 74 days, with peaks around
60 and 90 days. Similarly, the share of trade receivables represents 25% of total
assets with an average duration of 77 days. Different explanations are provided in
the literature to motivate such behavior. In addition to limited access to finance
from the banking sector (Petersen and Rajan 1995), suppliers have information
superiority over banks in lending to their customers (Buckart and Ellingsen 2004)
and they can monitor their financial position at lower costs (Jain 2001).
4. Econometric Analysis
Our empirical investigation is based on panel data regressions of leverage proxies
on firm attributes that different theories predict to be important in explaining
capital structure decisions. We first begin with a brief discussion of financial leverage
proxies adopted and then examine the list of relevant explanatory variables as
derived from theory.
4.1. Financial structure proxies
Following Rajan and Zingales (1995), the choice of leverage proxy depends on the
objective of the analysis. In our case, we use four different measures of leverage
based on book values of relevant financial variables.
The first is the ratio of debt to equity (Lev 1), which is typically used to assess
whether a given firm is using more debt or equity to finance its business. The second
is calculated as the ratio of total debt to total assets (Lev 2). These two leverage
proxies have similar information content since a firm’s equity is equal to its total
assets minus its total debt. However, they have different ranges. The first ratio
may vary between 0 (firm without debt) and infinity (firm without equity). The
second ratio lies between 0 and 1. The third leverage proxy is computed as the ratio
of long-term debt to permanent resourcesh (Lev 3). Finally, the fourth measure is
calculated as the ratio of short-term debt to total assets (Lev 4).
By investigating various dimensions of financial structure proxies, our purpose
is to deal with both debt-equity choice and debt maturity structure among nonlisted manufacturing firms. By doing so, we are able to test if factors that influence
short-term debt are similar or different from those that influence aggregate debt.
Figure 1 shows the distribution of the four proxies over the period 1998–2003
and Table 2 reports their descriptive statistics. The ratio of debt to equity varies
between a minimum value of 0.06 and a maximum value of roughly 10 with an
h Permanent
resources are defined as the sum of equity and long-term debt.
50
40
0
10
20
Percent
30
Percent
30
20
10
0
0
0
4
6
Measure 1: Debt to Equity ratio
8
1
10
0
0
.4
.6
Measure 2: Total debt to total assets
Fig. 1.
.8
.2
.4
.6
.8
Measure 4: Short term debt to total assets
.2
Distribution of financial structure proxies.
Source: Drawn by the author on the basis of the database.
.2
.4
.6
.8
Measure 3: Long term debt to long term fundings
2
40
30
Percent
20
10
0
40
30
Percent
20
10
1
1
262
0
January 2, 2010 13:56 WSPC/MEDJ
00012
L. Achy
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
Table 2.
Lev
Lev
Lev
Lev
1
2
3
4
263
Descriptive statistics of leverage proxies.
Mean
Standard
deviation
Min
Max
Skewness
Kurtosis
2.55
0.67
0.36
0.47
2.06
0.20
0.25
0.23
0.06
0.05
0.03
0.04
9.98
0.94
0.97
0.98
1.40
−0.50
0.66
0.05
4.57
2.69
2.42
2.40
Source: Author’s computation from the dataset. These are nonweighted statistics.
average of 2.55. The share of debt in total assets amounts to 0.67 on average and
ranges between 0.05 and 0.94. The average value of the third leverage ratio (longterm debt to permanent resources) is 0.36 with values located between 0.03 and
0.97. Finally, the ratio of short-term debt to total assets is 0.47 on average and
varies between 0.23 and 0.98. Financial leverage behavior among Moroccan firms
seems very heterogeneous, regardless of the proxy used for its measurement. This
finding is crucial as an important question in capital structure theory concerns the
extent to which firms’ financing decisions are driven by their own characteristics
rather than being the result of the institutional environment in which they operate
(Rajan and Zingales 1995).
Figure 1 shows that under a broadly similar macroeconomic and institutional
environment, there is substantial amount of variation in firms’ capital structure.
This finding justifies the emphasis we are putting on firms’ attributes as derived
from theory to account for leverage behavior of the Moroccan firms.
4.2. Potential determinants financial structure
We follow the empirical literature and focus on the most important firm attributes
derived from theory such as asset tangibility, firm size, growth prospects and profitability. The theoretical motivations behind the introduction of these attributes as
well as the proxies used to capture them are presented in what follows.
4.2.1. Asset tangibility
The existence of asymmetric information may induce lenders to require material
guarantees as collateral (Myers 1977, Harris and Raviv 1990). The type of assets a
firm possesses can be considered an important factor in determining its debt-equity
ratio.
Asset tangibility can also be related to the notion of a firm’s financial distress,
which is defined as its inability to meet interest or principal obligations to creditors
and hence may be forced to declare bankruptcy or agree to restructure its financial
claims. In particular, the costs of financial distress depend on the nature of assets
that a firm owns (Shleifer and Vishny 1992). If a firm retains large investments in
land, equipment and other tangible assets, it can sell them if it gets into financial
distress. However, firms with intangible assets such as brand names, patents and
January 2, 2010 13:56 WSPC/MEDJ
264
00012
L. Achy
human capital will face higher financial distress costs as unhappy customers and
employees can leave or seek alternative suppliers. In addition, tangible assets offer
more security than current assets. Therefore, firms with assets that can be used as
collateral may be expected to issue more debt.
However, large holdings of tangible assets could also suggest that a firm already
owns a stable source of return which yields more internally generated funds and
discourages it from relying on debt. Hence, a negative relationship between leverage
and asset structure may not be ruled out.
The expected effect of asset tangibility is also expected to vary between longterm and short-term leverage behavior. In principle, the role of tangible assets is
much more important for long-term debt than for short-term debt.
4.2.2. Size
Rajan and Zingales (1995) and Fama and Jensen (1983) argue that large firms have
relatively low information asymmetry problems. From a financial distress perspective, larger firms are more diversified and therefore expected to go bankrupt less
often compared to smaller ones as proposed by Warner (1977). Hence, larger firms
are expected to rely more on external finance and less on their own equity and
retained earnings, which implies a positive relationship between size and leverage.
The same argument also implies that larger firms can more easily have access to
long-term debt.
However, the “control rights model” developed by Hart (1995 and 2001) suggests
that small firms, in which owners do not wish to cede control rights to outside
investors, would tend to prefer debt over equity.i Therefore, the relationship between
leverage and size may also be negative. However, if small firms cannot meet their
needs through long-term loans, it is likely that most of their debt will be of short
maturity.
Since there is no perfect measure for size, we suggest three different proxies that
capture various aspects of the size effect. These are total sales, total employment
and total assets. The purpose is to assess the robustness of our econometric results
with respect to various proxies.
4.2.3. Growth prospects
Firms with high growth opportunities are more likely to exhaust internal funds
and search for additional capital through borrowing. However, Myers (1977) argues
that firms with growth potential will tend to have lower leverage. The reason is
that growth opportunities can produce moral hazard effects and push firms to take
more risk. This may explain why firms with important growth opportunities may
be considered risky and face difficulties in raising debt on favorable terms. The
i Anecdotic
Morocco.
evidence suggests that this is the case among small and medium non-listed firms in
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
265
relationship between leverage and growth opportunities is expected to be negative.
Empirically, Rajan and Zingales (1995) and Booth et al. (2001) corroborate the
above theoretical prediction.
As a proxy for growth prospects, we consider two alternatives. First, the change
of future sales (one year ahead) using data drawn from the annual survey of manufacturing firms for which information on sales are available until 2006. Although
based on actual data on future sales, this proxy is likely to introduce a selection
bias since some firms with prior strong growth prospects may fail to achieve them.
To partly overcome this issue, we consider using the firm’s past growth in sales as
an indicator for its future prospects. It turns out from our data that both proxies
are highly correlated.j This is why we only report econometric results with the first
proxy.
We investigate the impact of growth prospects on both the short and long-term
leverage behavior of the Moroccan manufacturing firms.
4.2.4. Profitability
The pecking order theory, developed by Myers and Majluf (1984) and Myers (1984)
suggests that firms prefer raising capital, first from retained earnings, second from
debt, and third from issuing new equity. According to that theory, more profitable
firms have more internal financing available. Therefore, we should expect a negative
relationship between leverage and profitability. However, it is also likely that some
firms use their profitability as an argument for additional external finance. It seems
that both explanations may be uncovered by making the distinction between longterm and short-term leverage. We use the ratio of operating income to total assets
as an indicator for profitability.
4.2.5. Other potential explanatory variables
There are other firm-specific variables that are potentially expected to be relevant in
understanding the financial leverage behavior of Moroccan firms, such as the degree
of concentration of ownership, the separation of ownership from control, and the
share of foreign ownership. Unfortunately, data on these variables are either absent
from our dataset or available only for a single year. This is why we were unable to
include them in our econometric analysis.
4.3. Econometric results
This section presents the results of panel data regressions of various debt measures
against the set of explanatory variables listed earlier. We run both random and
fixed effects regressions and opt for the later on the basis of Hausman test. The
estimates are reported in Table 3.
j Data on sales are highly auto correlated which means that past sales is good proxy for future
sales. Using the first or the second proxy does not seem to statistically change our results.
−0.10∗∗
−0.11∗∗∗
−0.12∗∗∗
(3)
−0.06∗∗
(−2 .90 )
−0 .06 ∗
(−1 .85 )
(−3 .67 ) (−2 .83 ) (−3 .76 )
(2)
(1)
−0.02∗∗∗
(−3 .71 )
(−1 .86 )
−0.05∗
(4)
−0 .02 ∗
(−1 .62 )
(−1 .66 )
−0.05∗
(5)
0.08
(0 .53 )
0.13∗∗
(2 .12 )
2859
0.12
5.91
0.00
−0.03∗∗
(−3.98)
0.03∗∗
(2 .77 )
−0.23∗∗∗
(−5 .36 )
2863
0.14
8.19
0.00
−0.03∗∗∗
(−3 .86 )
(1 .75 )
0.07∗
(7)
2380
0.10
5.84
0.00
0.03∗
(1 .78 )
0.12∗∗
(2 .50 )
−0 .09
(−0 .69 )
0.07
(1 .47 )
(8)
0.06
(0 .99 )
(−4 .29 )
−0.13∗∗∗
(10)
0 .02
(0 .22 )
(−4 .23 )
−0.16∗∗∗
(11)
−0.10∗∗∗
(−2 .94 )
(12)
Short-term debt to total assets
2863
0.14
5.67
0.00
2859
0.25
7.18
0.00
2380
0.23
6.58
0.00
2863
0.19
7.23
0.00
−0.1∗∗∗
0.06∗∗∗
(−9.60)
(7.54)
0.01
0.04∗∗∗
0.04∗∗∗
0.04∗∗∗
(0 .20 )
(3 .56 )
(3 .41 )
(3.92)
0.11∗∗ −0.41∗∗∗ −0.37∗∗∗ −0.31∗∗∗
(2 .42 ) (−8 .44 ) (−6 .63 ) (−6 .35 )
0.02
(0 .48 )
(9)
Long-term debt to long
terms funds
term debt
term debt
( LongLong
) and finally ( Short
). The estimation is based on panel data fixed effects method. The fixed effects are with respect to
term debt−Equity
Total assets
industries. The industries in the database are classified as follows: agro-food industry, garment industry, textiles, chemicals, electric and electronic
industries, and other industries. t- Statistics are reported in parentheses. * Significance levels at 10%, ** at 5% and *** at 1%.
debt
Total debt
Note: Author’s estimation on the basis of data from FACS and ICA surveys. Four dependant variables are considered, ( Total
), ( Total
),
Equity
assets
log (assets)
(−2 .04 )
−0.05∗∗
(6)
Debt to total assets
−0.09∗∗∗
(−3.50)
Growth potential
0.11∗∗∗ 0.04
0.10∗∗∗
0.03∗∗
0.08
(3 .18 ) (0 .96 ) (2 .93 )
(3 .06 )
(0 .80 )
Profitability proxy −0.15
−0.33∗∗ −0.26∗
−0.19∗∗∗ −0.25∗∗∗
(−1 .01 ) (−2 .08 ) (−1 .74 ) (−4 .68 ) (−5 .31 )
Number of
observations
2745
2278
2479
2859
2380
R2 (overall)
0.19
0.17
0.14
0.17
0.19
F -statistic
8.58
8.67
8.19
8.57
7.99
p-value
0.00
0.00
0.00
0.00
0.00
log (employment)
Size proxy
log (sales)
Tangible assets
Explanatory
variables
Debt equity ratio
Dependant variables
Regression results.
266
Table 3.
January 2, 2010 13:56 WSPC/MEDJ
00012
L. Achy
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
267
4.3.1. Debt-equity issue
The debt-equity issue is dealt with by using two complementary leverage proxies.
The first is the ratio of total debt to total equity. The second is the share of total
debt in total assets. As indicated earlier, these two leverage proxies have similar
information content; hence their results are broadly similar. This is why we focus
our comments on regressions (4), (5) and (6) reported in Table 3 under the second
leverage proxy.
First, there seems to be a negative relationship between the share of tangible
assets and leverage. The estimates reveal that the negative relationship is robust
to the size proxy used, with statistical significance at 95% confidence level when
size is measured by sales and at 90% confidence level when size is proxied either
by employment or assets. These findings are inconsistent with those of Titman
and Wessels (1988) as well as with those of Ozkan (2000) who found a positive
relationship between the importance of tangible assets in total assets and total
debt ratios. However, our results, although they may seem counterintuitive, are not
completely surprising. Theoretically, as mentioned earlier, firms with large amounts
of tangible assets probably already own a stable source of return that pushes them
to resort to internal funds rather than debt. From an empirical perspective, Booth
et al. (2001) have also found a negative relationship between tangibility of assets
and total debt ratios of listed firms in eight out of the 10 developing countries
covered in their study.k Therefore, the magnitude of the aggregate debt may not
be necessarily constrained by a narrow collateral basis, as captured by the share of
tangible assets in total assets.
Second, the relationship between size and leverage appears to be statistically
significant but with a negative sign. Our estimates indicate that the direction of
this relationship is not affected by the proxy selected to capture firms’ size. Therefore, small manufacturing firms are relatively more indebted compared to larger
firms. Our estimates are inconsistent with those found by Rajan and Zingales
(1995) for OECD countries except for Germany. They are also not in line with
those of Booth et al. (2001) except for two countries (Turkey and Zimbabwe) out
of ten.
The negative relationship we found corroborates the “control rights model”
developed by Hart (1995, 2001), which suggests that owners of small firms tend
to increase their debt instead of strengthening their capital by opening it to outside investors. The fundamental objective of this category of owners is to keep
control over their business. As a substantial share of small and medium Moroccan manufacturing firms is family owned with a high concentration of capital, the
negative relationship found between size and indebtedness does not seem to be
puzzling.
kA
negative relationship has been found in eight cases: Brazil, India, South Korea, Jordan,
Malaysia, Pakistan, Turkey and Thailand. The relationship was positive only in two cases: Mexico
and Zimbabwe.
January 2, 2010 13:56 WSPC/MEDJ
268
00012
L. Achy
Growth potential is another relevant explanatory variable of financial structure
of Moroccan manufacturing firms. Our estimates show that growth potential is positively associated to leverage. The relationship appears to be statistically significant
in two cases out of three. Our estimates are in line with the hypothesis according to
which firms with promising growth prospects tend to exhaust their internal funds
and to resort more intensively to debt.
As far as profitability is concerned, its relationship with leverage turns out to be
negative and statistically significant regardless of the size proxy used. The negative
relationship reveals that firms generating high returns from their businesses are
likely to maintain low levels of debt. This finding provides evidence supporting
“the pecking order theory” suggested by Myers and Majluf (1984) that firms prefer
internal funding and turn to external resources as a secondary option. It should
be emphasized that Rajan and Zingales (1995), and Booth et al. (2001) found
similar results respectively for OECD countries and listed companies in developed
countries.
4.3.2. Debt maturity issue
An examination of Fig. 1 suggests that there are marked differences between the
distributions of long-term debt ratio as compared to short-term debt ratio in the
Moroccan manufacturing firms. In addition, while Moroccan firms tend to be characterized by relatively high total debt ratios, their long-term debt ratios are rather
low. Both findings motivate our interest in dealing with the debt maturity issue by
separating the long-term component from the short-term component of debt.
The share of long-term debt in permanent funding is used as proxy for the longterm dimension of leverage. The corresponding econometric estimates are presented
in regressions (7), (8) and (9) in Table 3. Unlike our previous results, there seems
to be a positive relationship between the tangibility of firms’ assets and the degree
of their reliance on long-term debt. This relationship is, however, not statistically
significant except when total sales is used as proxy for size with a modest 90%
confidence level.
It is worth mentioning that unlike short-term debt, long-term debt originates
exclusively from the banking sector. Two competing explanations may lie behind
the absence of a clear-cut relation between tangibility of assets and access to longterm funds. The first is demand driven. The availability of collateral is a necessary
condition to have access to long-term funds. But having collateral may not automatically result in higher demand for long-term debt. The second is supply driven.
The availability of collateral is a necessary but not sufficient condition to have
access to long-term debt from the banking sector.
If we consider that in making their financial decisions, firms attempt to match
the duration of their liabilities with the liquidity of their assets, then the demand
driven explanation seems more plausible. When examining their balance sheets,
manufacturing firms appear to be endowed with more long-term funds to cover
their illiquid assets than what is suggested by the static matching rule. Long-term
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
269
funds (total equity and long-term debt) represent 51% of total assets compared to
33% for illiquid assets.
So how can we interpret the negative relationship found earlier when the aggregate debt ratio was used as a proxy for leverage? The negative relationship was
driven by the over-representation of short-term debt in the debt portfolio of Moroccan firms. From a managerial point of view, it makes sense for a firm that carries a
large share of its assets in short-term inventories and accounts receivable to accumulate a higher proportion of short-term debt. Therefore, there is a strong consistency
between the results uncovered so far, and further confirmed in regressions (10), (11)
and (12) on the relationship between leverage proxy and asset tangibility.
The relationship between long-term debt ratio and size lends support to the
previous finding revealed when the total debt ratio is used as the leverage proxy.
This relationship comes out with a negative and statistically significant sign in two
cases out of three. In particular, firms with large levels of sales or assets tend to
raise less long-term debt. The most plausible explanation is probably that larger
firms rely much more on their retained earnings for their long-term financial needs
in comparison with small and medium firms. This hypothesis seems to be confirmed
when looking at the share of retained earnings in total assets for various size categories. Retained earnings represent 12% of total assets for the whole sample during
the period 1998–2003, 16% for large firms and 6.5% for small and medium firms.l
These findings provide support to the pecking order theory in shaping financial
decisions of the Moroccan manufacturing firms.
The relationship between growth potential and the short-term debt ratio
appears to be positive and highly significant as shown in the last three regressions of Table 3. As growth potential is measured by variation in annual sales,
higher sales require, ceteris paribus, more inventories, receivables and cash. Growth
potential in our case reflects short-term opportunities that are subject to cyclical
shifts, it seems rational to expect that short-term debt will adjust to meet the
extra liquidity needs. From this perspective, growth potential would be irrelevant
in driving long-term debt ratio, which corroborates our previous findings.
Finally, profitability emerges as a factor that exerts a positive and statistically
significant effect on the long-term debt ratio as reported in regressions (7), (8) and
(9). In addition, this result is robust to the size proxy selected. Conversely, the
impact of profitability on short-term leverage is negative. These results indicate
that profitability induces the leverage behavior of firms in two directions. On one
hand, it leads firms, in the short-term, to switch partly from their dependence on
short-term debt, probably by retaining a portion of their internally generated cash
flows. On the other hand, profitable firms seem to use their “good accounts”, as
suggested by Ross (1977), to signal their quality to potential borrowers and have
more easy access to long run debt.
l For
the purpose of this classification of firms, we used the Moroccan official definition for which
a large firm has total annual sales over 75 million DH (approximately US$9 million).
January 2, 2010 13:56 WSPC/MEDJ
270
00012
L. Achy
Table 4.
The impact of firms’ attributes on leverage proxies.
Total debt to total assets Long-term debt ratio Short-term debt ratio
Asset tangibility
Size
Negative and fairly robust
Negative and fairly
robust
Growth opportunities
Positive and fairly
robust
Profitability
Negative and strong
Not clear cut
Negative and fairly
robust
Not significant
Negative and strong
Positive but weak
Positive and strong
Negative and strong
Positive and strong
Note: Constructed on the basis on the econometric estimates.
To sum up, our findings from dealing with the debt equity choice and the debt
maturity choice are summarized in Table 4.
5. Conclusions and Policy Implications
The objective of the paper is to provide empirical evidence on the behavior of
leverage among non-listed firms in the context of the MENA region. The main
findings of the paper are as follows.
Our descriptive analysis of Moroccan firms’ sources of funding reveals their
potential financial fragility through their excessive reliance on short-term external finance. However, one of the basic rules in financial management is that firms
need to match the maturity of their debt with the degree of liquidity of their
assets. At the aggregate level as well as by size class, there seems to be no special
worry about the financial structure of Moroccan firms. Their long-term sources of
finance cover their fixed assets and their current assets are larger than their current
liabilities.
Moroccan firms are also characterized by their large proportion of credit to
suppliers and short-term creditors. Understanding the determinants of this specific
category of credit and its interactions with access to and cost of banking credit are
very important issues that are beyond the purpose of this paper. They are left for
future research.
Our empirical investigation is based on panel data regressions of leverage proxies
on firm attributes such as asset tangibility, size, expected growth and profitability.
For the aggregate leverage proxy, the key results can be summarized as follows.
First, a negative and statistically significant relationship emerges between the
share of tangible assets and leverage. Second, the relationship between size and
aggregate leverage is negative suggesting that small firms increase their debt instead
of opening their capital. This explanation fits with the “control rights model”
developed by Hart (1995, 2001). The challenge for policy makers is to provide
an environment in which individual and family owned firms can retain sufficient
profits in their businesses. Yet, the tax regime in Morocco has in 2008 removed
tax rebates granted before to firms retaining their profits. The new regime stipulates that the same corporate tax is applied regardless of the firm’s profit allocation.
Based on our results, we believe that there is scope for fiscal policy that will provide
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
271
incentives to retain profits and encourage investment in growth oriented strategies.
Third, growth potential is positively associated to leverage. This is in line with
the hypothesis according to which firms with promising growth prospects tend to
exhaust their internal funds and to resort more intensively to debt. Fourth, the
relationship between profitability and aggregate leverage turns out to be negative
and statistically significant in most cases. This finding fits with “the pecking order
theory”.
We then decompose leverage into its short and long-term dimensions in order to
identify similarities and differences in the impact of various determinants of firms’
behavior. Fairly interesting results emerge from this analysis and provide further
understanding to time dimension of leverage choices in Moroccan manufacturing
firms.
First, unlike our previous results, there seems to be a positive but statistically
weak relationship between the tangibility of firms’ assets and the degree of their
reliance on long-term debt. Two competing explanations may lie behind the absence
of a clear-cut relation between tangibility of assets and access to long-term funds.
The first is demand driven. The second is supply driven. If we consider that in
making their financial decisions, firms attempt to match the duration of their liabilities with the liquidity of their assets, then the demand driven explanation seems
more plausible. It implies that availability of collateral is a necessary condition to
have access to long-term funds. But having collateral may not automatically result
in higher demand for long-term-debt. Moreover, manufacturing firms appear to be
endowed with relatively more long-term funds than with what they really need
for covering their illiquid assets. The relationship between asset tangibility and
short-term leverage maintains a negative sign. From a managerial point of view, it
makes sense for a firm that carries a large share of its assets in short-term inventories and accounts receivable to accumulate a higher proportion of short-term debt.
Therefore, there is a strong consistency between the results uncovered for different
measures of leverage and asset tangibility.
Second, the relationship between long-term debt ratio and size lends support
to the previous finding uncovered when the total debt ratio is used as the leverage
proxy. In particular, firms with large levels of sales or assets tend to raise less longterm debt. The most plausible explanation is probably that larger firms rely much
more on their retained earnings for their long-term financial needs in comparison
to small and medium firms. This hypothesis seems to be confirmed when looking
at the share of retained earnings in total assets for various size categories. These
findings provide support to the pecking order theory in shaping financial decisions
of Moroccan manufacturing firms.
Third, the relationship between growth potential and short-term debt ratio
appears to be positive and highly significant. As growth potential is measured
by variation in annual sales, higher sales require, ceteris paribus, more inventories,
receivables and cash. Since growth potential in our case reflects short-term opportunities that are subject to cyclical shifts, it seems rational to expect that short-term
January 2, 2010 13:56 WSPC/MEDJ
272
00012
L. Achy
debt will adjust to meet the extra liquidity needs. From this perspective, growth
potential would be irrelevant in driving long-term debt ratio.
Fourth, profitability emerges as a factor that exerts a positive and statistically
significant effect on the long-term debt ratio. In addition, this result is robust to the
size proxy selected. Conversely, the impact of profitability on short-term leverage
is negative. These results indicate that profitability induces the leverage behavior
of firms in two opposite directions. On one hand, it leads firms, in the short-term,
to switch partly from their dependence on short-term debt, probably, by retaining
a portion of their internally generated cash flows. On the other hand, profitable
firms seem to use their “good accounts”, as suggested by Ross (1977), to signal
their quality to potential borrowers and have more easy access to long run debt.
Finally, while the existence of supply-side financial constraints tends to be the
main anecdotal evidence usually referred to in explaining financial structure of
Moroccan firms, the evidence emerging from our paper indicates that a substantial
amount of the explanation is demand driven. Hence, the barriers to firms’ growth
frequently perceived as financial, are often managerial and cultural. The availability
of external funds alone may not be sufficient to solve the problem.
References
Agarwal, S. and H. Mohtadi. 2004. “Financial markets and the financing choice of firms:
Evidence from developing countries.” Global Finance Journal, 15(1): 57–70.
Aggarwal, R. and S. Jamdee. 2003. “Determinants of capital structure: Evidence from G-7
countries.” Kent University working paper.
Ayyagari, M., A. Demirguç-Kunt and V. Maksimovic. 2005. “How important are financing
constraints? The role of finance in the business environment.” World Bank Mimeo.
Beck, T. and A. Demirguç-Kunt. 2006. “Small and medium-size enterprises: Access to
finance as a growth constraint.” Journal of Banking and Finance, 30(11): 2931–2943.
Beck, T., A. Demirguç-Kunt and V. Maksimovic. 2008. “Financing patterns around the
world: Are small firms different?” Journal of Financial Economics, 89(3): 467–487.
Beck, T., A. Demirguç-Kunt and V. Maksimovic. 2005. “Financial and legal constraints
to firm growth: Does firm size matter?” Journal of Finance, 60(1): 137–177.
Booth, L., V. Aivazian, A. Demirguc-Kunt and M. Vojislav. 2001. “Capital structures in
developing countries.” Journal of Finance, 56(1): 87–130.
Bukart, M. and T. Ellingsen. 2004. “In-kind finance: A theory of trade credit.” American
Economic Review, 94(3): 569–590.
Cobham, D. and R. Subramaniam. 1998. “Corporate finance in developing countries: New
evidence for India.” World Development, 26(6): 1033–1047.
Danielson, M. and J. Scott. 2004. “Bank loan availability and trade credit demand.”
Financial Review, 39(4): 579–600.
Fama, E. F. and M. Jensen. 1983. “Agency problem and residual claims,” Journal of Law
and Economics, 26(2): 327–349.
Glen, J. and B. Pinto. 1994. “Debt and equity: How firms in developing countries choose?”
Discussion paper no. 22, International Finance Corporation.
Harris, M. and A. Raviv. 1990. “Capital structure and the informational role of debt.”
Journal of Finance, 45(2): 321–349.
Hart, O. 1995. Firms, Contracts, and Financial Structure, Oxford University Press.
January 2, 2010 13:56 WSPC/MEDJ
00012
Corporate Capital Structure Choices in MENA
273
Hart, O. 2001. “Financial contracting.” Journal of Economic Literature, 39(4): 1079–1100.
Holmes, S. and P. Kent. 1991. “An empirical analysis of the financial structure of small
and large Australian manufacturing enterprises.” Journal of Small Business Finance,
1(2): 141–154.
Jain, N. 2001. “Monitoring costs and trade credit.” Quarterly Review of Economics and
Finance, 41(1): 89–100.
Jensen, M. and W. Meckling. 1976. “Theory of the firm: Managerial behavior, agency
costs and ownership structure.” Journal of Financial Economics, 3(4): 305–360.
Modigliani, F. and M. Miller. 1958. “The cost of capital, corporation finance and the
theory of investment.” American Economic Review, 48(3): 261–207.
Myers S. and N. Majluf. 1984. “Corporate financing and investment decisions when firms
have information that investors do not have.” Journal of Financial Economics, 13(2):
187–221.
Myers, S. 1984. “The capital structure puzzle.” Journal of Finance, 39(3): 575–592.
Myers, S. 2001. “Capital structure.” Journal of Economic Perspectives, 15(2): 81–102.
Myers, S. 1977. “Determinants of corporate borrowing.” Journal of Financial Economics,
5(2): 47–175.
Ozkan, A. 2000. “An empirical analysis of corporate debt maturity structure.” European
Financial Management, 6(2): 197–212.
Petersen, M. and R. Rajan. 1997. “Trade credit: Theories and evidence.” Review of Financial Studies, 10(3): 661–692.
Rajan, R. and L. Zingales. 1995. “What do we know about capital structure? Some evidence from international data.” Journal of Finance, 50(5): 1421–1460.
Ross, G. C. 1977. “The determination of financial structure: The incentive signaling
approach.” Bell Journal of Economics & Management Science, 8(1): 23–40.
Shleifer A. and R. Vishny. 1992. “Liquidation values and debt capacity: A market equilibrium approach.” Journal of Finance, 47(4): 1343–1366.
Titman, S. and R. Wessels. 1988. “The determinants of capital structure choice.” Journal
of Finance, 43(1): 1–19.
Warner, J. B. 1977. “Bankruptcy costs: Some evidence.” Journal of Finance, 32(2):
337–348.